|Publication number||US4215414 A|
|Application number||US 05/884,325|
|Publication date||Jul 29, 1980|
|Filing date||Mar 7, 1978|
|Priority date||Mar 7, 1978|
|Also published as||DE2907992A1|
|Publication number||05884325, 884325, US 4215414 A, US 4215414A, US-A-4215414, US4215414 A, US4215414A|
|Inventors||Kenneth A. Huelsman|
|Original Assignee||Hughes Aircraft Company|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (19), Referenced by (34), Classifications (18), Legal Events (2)|
|External Links: USPTO, USPTO Assignment, Espacenet|
This invention relates to a digital system for video display, and more particularly to a method and apparatus for digital video output processing to approximate the desirable smooth picture characteristics of an analog video output such as is generated by a TV camera or scan converter tube.
There are many reasons for using a digital system for video display. One is to provide a more versatile video scan converter than is possible with an analog scan converter tube. However, a digital video system does have its own problems. A primary problem is that the quantified picture elements (pixels) present a quantified or structured appearance of the display.
What is required in a digital system for video display is video output processing to transform the structured output to an output that emulates the gaussian characteristics of an analog output. The transformed digital output may then be displayed with the picture characteristic of an analog video system. An advantage is not only reduction of the quantization effects characteristic of a digital display system, but also a reduction of interfield flicker typical of digital display systems.
In the past, the digital output transformation has usually depended upon analog filtering to smooth the quantified pixels. A problem with that technique is that the smoothing occurs only along the horizontal axis. Also analog filters involve phase shifts and resultant rise time characteristics which do not match the desired gaussian function. In order to transform the digital video output to an output which emulates an analog video output, filtering would be required in both axes, and that simply would not be possible with just a filter to transform the digital video output into an analog video output.
In accordance with the present invention, a pseudogaussian distribution is provided in the digital video output for a single video event in both display axes in order that when displaying any one line, the information from at least immediately adjacent lines is also partially imaged, and within any one line, the information from similarly processed pixels of the same line will also be partially imaged utilizing at least immediately adjacent pixels. This partial imaging of surrounding each pixel of a line displayed, emulates the gaussian characteristics of an analog system. The partial imaging technique is carried out for a given line L by reading at least line L+1 out of digital memory, and preferably delaying that line one line scanning interval while line L is read through a demultiplexer in order to present at the output of the delay means a video data line L-1 to add pixel by pixel to line L+1, with a predetermined weighting factor W less than unity (typically 0.5). Following that, the resulting weighted sums for the pixels of the line L are transmitted through two cascaded pixel delay elements in order to make adjacent pixels P-1 and P+1 available for addition to each pixel P, again with a predetermined weighting factor W less than unity (typically 0.5), with the result that each pixel value has a contribution from not only adjacent pixels in the vertical and horizontal axis, but also a contribution from diagonally adjacent pixels for an approximation of gaussian distribution all around each pixel on a pixel by pixel basis while display takes place. For a more simple, but effective implementation, the line delay means used for presentation of line L-1 could be omitted. Lines L and L+1 would be read out of memory using a demultiplexer as before. This simple implementation would effect unidirectional vertical filtering and bidirectional horizontal filtering that, in practice, approaches the effectiveness of the full L-1, L and L+1 line filtering. Alternatively, all three lines L-1, L and L+1 may be read and brought into coincidence, with suitable delay means possibly even with half pixel skewing of line L in respect to line L-1 and L+1 for vertical filtering of two pairs of pixels diametrically opposite each other instead of one pair. Full filtering could also be achieved without the line delay means by using a demultiplexer to read pixels from all three lines, and forming the weighted sum of corresponding pixels of all three lines directly.
The novel features that are considered characteristic of this invention are set forth with particularity in the appended claims. The invention will best be understood from the following description when read in connection with the accompanying drawings.
FIG. 1 is a block diagram illustrating an exemplary environment for the invention.
FIG. 2 illustrates a typical video signal.
FIG. 3 illustrates the video signal of FIG. 2 quantized and analog filtered.
FIG. 4 illustrates the quantized signal of FIG. 3 subjected to digital filtering.
FIGS. 5, 6 and 7 are graphics useful in understanding the objectives of the invention.
FIG. 8 is a block diagram illustrating the general organization and operation of the invention.
FIG. 9 illustrates a preferred embodiment.
FIG. 10 is a timing diagram for FIG. 9.
FIG. 11 illustrates a variant of the embodiment of FIG. 9.
Before describing a preferred embodiment, the purpose for and advantages of the invention will first be further described with reference to FIG. 1. The ultimate purpose is to approximate the desirable smooth picture characteristic of an analog display, such as generated by an ultrasound system 10 of the type referred to in U.S. Pat. No. 3,864,660, where the X, Y and θinformation provide coordinate data for the analog video signal Z. This information is used by a digitizing and scan converting unit 12 to convert the video signal Z to digital form (Z data) in a coordinate system referred to a video display unit with X and Y addresses for each picture element (pixel). The X and Y addresses are used to store the Z (video) data in a main digital memory 16 from which the data is read for display, one line at a time, in alternate fields of odd and even lines. A timing and control unit 18 synchronizes operation of the digitizing and scan converting unit 12 and the main memory 16 while storing video data. A second timing and control unit 20 synchronizes operation of the main memory with the video display unit while reading out video data.
The video data could be displayed directly out of the memory 16, but each pixel is quantized to one of 2n values, where n is the number of binary digits (bits) in the Z data. Therefore, the output video from the ultrasound system, or equivalent source (e.g., TV camera) shown in FIG. 2, is quantized. FIG. 3 shows a typical quantized video signal. Each pixel is stored as an n-bit number representing the quantized level of the video signal at the time the pixel sample is taken. If the quantized signal of FIG. 3 were to be displayed, the display would have a quantified or structured appearance, whereas the actual video output of events is generally gaussian. This can be better appreciated by considering a single impulse or video event which, when quantized, produces a squarewave video event or pulse. If displayed as a squarewave over several pixels and lines, the result is a rather "hard" image event with a well defined edge and uniform gray scale. The problem then is that this does not display the event as one would expect it to be displayed for the reason that an analog source of a single video event is generally gaussian.
The important feature of analog video sources is that for a single impulse or squarewave video event, the analog video signal has essentially equal skirts on each side of the main video pulse due to the inherent gaussian characteristics of the analog signal source. Therefore the quantized display shown in FIG. 3 is not a very faithful reproduction of the analog signal. It would be helpful to use an analog filter of the video output to the display system. The result would then be as represented by the dotted line in FIG. 3, but that is still a distortion of the true analog video of FIG. 2. It would be preferable to use digital filtering to produce a more faithful quantized reproduction of the analog video, as shown in FIG. 4, by providing a single video event with skirts characteristic of an analog signal source. This will be more explicitly described with reference to FIGS. 5, 6 and 7.
In FIG. 5 there is shown a small square representing a single video event which is scanned horizontally. When scanned, the analog system (e.g. TV camera) will have some response in line L-1, and L+1, as well as line L, with the largest response in line L as shown in FIG. 6. This results when scanning any one line due to the adjacent lines picking up the skirts of the impulse event. Similar responses occur in line L for pixels intervals before and after the event. This is because the skirt of the impulse rises slowly and falls slowly compared to the pulse width (scan interval). When processed in analog form, the event will be displayed as a smooth flickerless spot, but if sampled and quantized, for processing in digital form, the lines before and after (and the pixels before and after) will be zero, resulting in a "hard" display of the event.
To improve the display, of a system using digital processing of the video, such as for scan conversion, the present invention provides digital filtering in both the horizontal and vertical dimension to provide adjacent pixels in lines L-1 and L+1 (and adjacent pixels in columns P-1 and P+1) with some amplitude approximating the gaussian characteristics of the signal source. The result is lower flicker in the display, giving the viewer a softer (smoother) image which is more like smooth flickerless display achieved using analog processing. This digital filtering in both the horizontal and vertical directions is provided by a digital video output processing unit 22 (FIG. 1) such that for a 3 by 3 matrix of pixels as shown in FIG. 7, the center pixel E (line L, pixel P) will have a value dependent on the weighting, W, used in adding pixels in the vertical and horizontal directions as follows:
PL =E+W(B+D+F+H)+W2 (A +C+G+J)
where the weighting factor is less than unity. Assuming W=0.50 the value of the pixel will be:
PL =E+0.5(B+D+F+H)+0.25 (A+C+G+J)
Assuming all squares A through J have a full scale value of 1, then the value of the pixel in the center is:
PL =1+0.5 (4)+0.25 (4)=4
This indicates a 4 times expansion of gray scale range required of the digital-to-analog converter for the video display unit. Changing the weighting factor to 0.25 for adjacent pixels results in a pixel value of:
Again assuming a full scale value of 1, the pixel value is:
This value is considered realistic in terms of both filtering capability (horizontally and vertically) and flicker reduction from an interlaced line display system. The amount of "gaussian" value to be added in can be adjusted either by shifting down the adders bits in a digital adder, or by performing a more elaborate multiplication (or division by the reciprocal of the weighting factor). It would also be possible to use separate digital-to-analog converters for B, D, F, H and A, C, G, J followed by analog summing. Further improvement could be achieved by skewing line L to the left, for example, relative to lines L-1 and L+1 by a half pixel, and digital filtering with all adjacent pixels in an analogous manner. Still further improvement could be achieved by processing for pseudogaussian characteristics over more than one line on each side, and over more than one pixel similarly processed on each side.
Before proceeding with a description of a preferred embodiment, the organization of the invention will first be described with reference to FIG. 8 wherein a main memory 30 for digitized video output of a system is shown connected to buffer delay lines 32 which receive the video output and present three lines at a time to a pair of adders 34 and 36 with the following pattern:
______________________________________ L-1 L L+1______________________________________ 0 1 2 2 3 4 . . . . . .ODD . . .FIELD 14 15 16 16 17 18 . . . . . . . . . 1 2 3 3 4 5 . . .EVEN . . .FIELD . . . 13 14 15 15 16 17______________________________________
Lines L-1 and L+1 are added through a binary adder 34 and the output divided by 2n is added to line L through a binary adder 36, one N-bit pixel at a time. All pixel data is processed in parallel. Therefore, all digital data lines are to be understood to be comprised of N bit lines, with N typically equal to 4. If the output of the adder 34 is to be weighted by 0.25, for example, that output must be divided by 22 =4. That is easily accomplished by adding the N-2 most significant bits to the N-2 least significant bits of the pixel value in line L. For instance, assuming 4-bit pixels, the maximum number of bits in a sum will be 5. By adding the three most significant bits to the three least significant bits of the 4-bit pixel value of line L, the sum is effectively divided by four, which is a multiplication by the weighting factor 0.25. The vertically filtered video output of the adder 36 is then applied to a pair of adders 38 and 40 through pixel delay elements 42 which receive a line of video data, one pixel at a time and present it to the adders with the following pattern:
______________________________________P-1 P P+1______________________________________0 1 2. . .. . .. . .13 14 1514 15 1615 16 17. . .. . .. . .______________________________________
The adder 38 sums the values of pixels P-1 and P+1, and the adder 40 adds the sum (divided by the reciprocal of the weighting factor) to the value of pixel P to develop the pixel value transmitted to a digital-to-analog converter for display.
It should be noted that for vertical filtering, the lines to be operated on are the odd (or even) lines with the adjacent lines, which requires reading at least line L and line L+1, and storing line L+1 through one extra line delay for use during the next line filtering cycle. This requires reading each line L directly into the adder 36 while delaying it through one delay line before presenting it to the adder 34 and using another delay line to connect the one delay line to the other input of the adder 34. The problem with that approach is that the first two lines of vertically filtered data will not be properly filtered. It would be better to read lines L and L+1 in a multiplexed manner, and to demultiplex them into the adders 34 and 36, with only line L+1 being stored in a delay line for use during the next line scanning cycle to present the line L-1 to the adder 36. Just the first line of every odd and even frame would then be only partially filtered, but that is not entirely objectionable. In fact, a simple but effective implementation would eliminate the one display line delay and use only lines L and L+1. This results in unidirectional vertical filtering which in actual practice approaches the effectiveness of the full filter. Still another alternative would be to multiplex the reading of all three lines L-1, L and L+1, and to demultiplex them for presentation to the adders. The result would be full bidirectional vertical filtering from the very first line without a delay line. However, the preferred embodiment to be described with reference to FIG. 9 is to read lines L and L+1 in a multiplexed manner, and to employ a delay line to store line L+1 for use during the next line scan cycle as line L-1.
The situation for horizontal filtering is not complicated by the need, or desire, to interlace for flicker free display. Consequently, the delay elements 42 of FIG. 8 may be implemented with just two delay elements, one 42a providing just one pixel delay to present the pixel P to the adder 40, and one 42b providing two pixel delays to present pixel P+1 to the adder 38. Pixel P-1 is applied directly to the adder 38 as shown. Since each pixel is a N-bit word presented in parallel, a delay of one pixel may be implemented with one bank of N flip-flops, and a delay of two pixels may be implemented with two banks of N flip-flops in cascade. The buffer delay lines 32 would similarly be implemented the necessary number of banks of flip-flops in cascade, each bank storing one pixel. The video data would be processed through these delays and adders in response to clock pulses in the usual manner well known to those skilled in the art of designing synchronous digital systems.
The preferred implementation for digital video output processing will now be described with reference to FIG. 9 wherein the main memory 43 is essentially the same as the main memory 16 of FIG. 1. The balance of FIG. 9 is then the digital video output processing unit 22 of FIG. 1 with a timing and control unit 44 providing the timing signals of FIG. 10 for synchronous control of the main memory and the digital video output processing unit.
The first timing signal out of unit 44 is a main memory read mode signal which enables the memory to read out two lines, pixel by pixel, in a multiplexed mode, first one line L for an address (Y ADDR) and then the next line L+1 of a frame. Note that the line L will be odd (or even) for interlaced line display and the next line will be even (or odd). The multiplexed data out of the main memory is demultiplexed by a demultiplexer 46 in response to a demultiplex (DEMUX) signal. The address for the pixels is controlled by a signal (X ADDR) that is periodically incremented to provide for reading out 16 pixels at a time in parallel. The demultiplexer thus provides the pixel data for lines L and L+1 on separate lines. The pixel data read is first converted from parallel to serial form by parallel-to-serial converters (P/S CONV) 48 and 50 and 52, but the bits of each pixel are handled in parallel. To accomplish demultiplexing, there are two load signals, LOAD A and LOAD B, generated for the parallel-to-serial converters 50 and 52 in order to be able to load one while shifting the other out, but only one load signal, LOAD (L+1) is generated for the parallel-to-serial converter 48. Thus, the converter 48 provides the demultiplexed line L+1 while the converters 50 and 52 together provide the demultiplexed line L. Two parallel-to-serial converters are required for the line L because pixels of a given line L in a converter must be held in overlapping time with pixels of the next line L in order to have the pixels of the given line L and of the line L+1 read out of the converters 48 and 50, or 52, synchronized.
The output (L+1) of the parallel-to-serial converter 48 is applied directly to an adder 54 and to a delay means 56 for providing one display line delay, such as a shift register having N stages for a line of pixels, each stage having n bits, i.e., by a shift register n bits wide. The output of the delay means 56 is applied to the adder 54 which adds lines L+1 and L-1. The least significant N-1 bits are applied to an adder 58 for addition to line L from converter 50, or 52, via multiplexer (MUX) 60, depending upon which of two gates G2 and G3 is enabled to transmit clock pulses to the parallel-to-serial converters (shift registers) 50 and 52. Note that the parallel-to-serial converter (shift register) 48 will receive clock pulses continuously. The AND gates G2 and G3 are alternately enabled by a signal SHIFT A/B, and its complement produced through an inverter 61. The signal SHIFT A/B is also applied to the multiplexer 60 to transmit the line L to the adder 58 from the converter 50 while it is receiving clock pulses from the gate G2, and from the converter 50 while it is receiving clock pulses. The signal SHIFT A/B thus not only gates clock pulses to the converters 50 and 52 but also connects the outputs of those converters to the adder alternately.
The serial pixel output of the adder 58 is applied to a first pixel delay element 62 (bank of N flips-flops) and then to a second pixel delay element 64 (bank of N flip-flops). That provides pixel P+1 at the input of the first pixel delay element and the pixels P and P-1 at the outputs of the first and second delay elements, respectively. Pixels P+1 and P-1 are added in an adder 66, and the sum (P+1)+(P-1) is added to pixel P in an adder 68. The output of the adder is applied to a digital-to-analog converter 70 to provide an analog video out signal synchronized by a line synchronizing signal (SYNC) which blanks the output of the digital-to-analog converter between line sweeps of a cathode ray tube (not shown).
As noted hereinbefore, unidirectional vertical filtering with bidirectional horizontal filtering could be, and has been, effectively used. To accomplish that, the delay line 56 and adder 54 are omitted so that the pixel values of line L+1 multiplied by a weighting factor W are added directly to the pixel values of the line L, as shown in FIG. 11. All of the rest of the system shown in the embodiment of FIG. 9 remains the same.
Although particular embodiments of the invention have been described and illustrated herein, it is recognized that modifications and variations may readily occur to those skilled in the art. For example, although filtering has been illustrated over a 3×3 array of pixels for each pixel of a line displayed (or a 3×2 array in the case of unidirectional vertical filtering), or the equivalent in the case of skewing line L relative to lines L-1 and L+1, filtering could also be achieved over a greater array of NXN pixels, where N may be 5, for example.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US3573789 *||Dec 13, 1968||Apr 6, 1971||Ibm||Method and apparatus for increasing image resolution|
|US3878536 *||Jul 10, 1972||Apr 15, 1975||Philips Corp||Apparatus for improving the shape of characters formed by a row and column coordinate matrix for display on a cathode-ray tube|
|US3921164 *||Jun 3, 1974||Nov 18, 1975||Sperry Rand Corp||Character generator for a high resolution dot matrix display|
|US3983320 *||Aug 25, 1975||Sep 28, 1976||Hughes Aircraft Company||Raster display histogram equalization|
|US3996421 *||Sep 2, 1975||Dec 7, 1976||Hughes Aircraft Company||Television display utilizing local area brightness control|
|US4023165 *||Jun 2, 1975||May 10, 1977||Rca Corporation||Signal modification techniques|
|US4032977 *||Jun 3, 1976||Jun 28, 1977||Xerox Corporation||Gray scale interpolation technique|
|US4038668 *||Oct 31, 1975||Jul 26, 1977||Honeywell Inc.||Apparatus for producing smooth and continuous graphic displays from intermittently sampled data|
|US4068266 *||Jun 3, 1976||Jan 10, 1978||Xerox Corporation||Statistical resolution conversion technique|
|US4127873 *||May 20, 1977||Nov 28, 1978||Rca Corporation||Image resolution enhancement method and apparatus|
|US4158838 *||Jan 16, 1978||Jun 19, 1979||Hughes Aircraft Company||In-raster symbol smoothing system|
|GB1193497A *||Title not available|
|GB1277013A *||Title not available|
|GB1299842A *||Title not available|
|GB1336079A *||Title not available|
|GB1393532A *||Title not available|
|GB1453589A *||Title not available|
|GB1510504A *||Title not available|
|GB1510978A *||Title not available|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US4280143 *||Sep 28, 1979||Jul 21, 1981||International Business Machines Corporation||Method and means for scale-changing an array of boolean coded points|
|US4323974 *||Mar 5, 1980||Apr 6, 1982||Ricoh Co., Ltd.||Method of restoring a picture through estimation|
|US4354186 *||Jan 21, 1980||Oct 12, 1982||U.S. Philips Corporation||Picture display device for displaying a binary signal generated by a picture signal generator as a binary interlaced television picture|
|US4384336 *||Aug 29, 1980||May 17, 1983||Polaroid Corporation||Method and apparatus for lightness imaging|
|US4393452 *||Jun 2, 1980||Jul 12, 1983||Ricoh Co., Ltd.||Method of reproducing a picture having improved resolution|
|US4454506 *||Sep 4, 1981||Jun 12, 1984||Bell Telephone Laboratories, Incorporated||Method and circuitry for reducing flicker in symbol displays|
|US4532503 *||Nov 8, 1982||Jul 30, 1985||International Business Machines Corporation||Sequence controlled pixel configuration|
|US4604614 *||Sep 26, 1983||Aug 5, 1986||International Business Machines Corp.||Video display system employing pulse stretching to compensate for image distortion|
|US4649568 *||Oct 22, 1984||Mar 10, 1987||Polaroid Corporation||Reconstitution of images|
|US4672369 *||Nov 7, 1983||Jun 9, 1987||Tektronix, Inc.||System and method for smoothing the lines and edges of an image on a raster-scan display|
|US4689675 *||May 23, 1985||Aug 25, 1987||Advanced Systems Development, Inc.||Digital scan converter and method therefor|
|US4796016 *||May 5, 1986||Jan 3, 1989||U.S. Philips Corporation||Pixel rounding method and circuit for use in a raster scan display device and a raster scan display device comprising such circuit|
|US4799105 *||Sep 16, 1987||Jan 17, 1989||International Business Machines Corporation||Modified technique for suppression of flicker in interlaced video images|
|US4833462 *||Aug 13, 1984||May 23, 1989||International Business Machines Corporation||Raster-scanned cathode ray tube display with cross-hair cursor|
|US4907152 *||Dec 5, 1988||Mar 6, 1990||The Boeing Company||Method of improving CT resolution|
|US4952921 *||Jun 9, 1988||Aug 28, 1990||Rockwell International Corporation||Graphic dot flare apparatus|
|US5001651 *||Nov 10, 1986||Mar 19, 1991||Auto-Trol Technology Corporation||Method and apparatus for interpolating groups of pixels on a scan line|
|US5005011 *||Dec 23, 1988||Apr 2, 1991||Apple Computer, Inc.||Vertical filtering apparatus for raster scanned display|
|US5083204 *||Oct 1, 1984||Jan 21, 1992||Hughes Aircraft Company||Signal processor for an imaging sensor system|
|US5132674 *||Jun 6, 1989||Jul 21, 1992||Rockwell International Corporation||Method and apparatus for drawing high quality lines on color matrix displays|
|US5315310 *||Dec 4, 1992||May 24, 1994||International Business Machines Corporation||Cathode ray tube display apparatus|
|US5440676 *||Jan 29, 1988||Aug 8, 1995||Tektronix, Inc.||Raster scan waveform display rasterizer with pixel intensity gradation|
|US5638310 *||Mar 21, 1995||Jun 10, 1997||Sgs-Thomson Microelectronics S.A.||Pixel matrix filter|
|US5754163 *||Aug 22, 1995||May 19, 1998||Lg Electronics Inc.||Liquid crystal display controlling apparatus|
|US6493036||Nov 17, 1999||Dec 10, 2002||Teralogic, Inc.||System and method for scaling real time video|
|US6545686||Feb 2, 1999||Apr 8, 2003||Oak Technology, Inc.||Cache memory and method for use in generating computer graphics texture|
|DE3134703A1 *||Sep 2, 1981||Apr 22, 1982||Sony Corp||Flimmern verhindernde schaltung|
|EP0146227A2 *||Oct 15, 1984||Jun 26, 1985||Tektronix, Inc.||System and method for raster-scan display|
|EP0566229A2 *||Feb 5, 1993||Oct 20, 1993||nVIEW CORPORATION||Method and apparatus for deinterlacing video inputs|
|EP0681280A2||Dec 8, 1989||Nov 8, 1995||Apple Computer, Inc.||Vertical filtering method for raster scanner display|
|EP0681281A2||Dec 8, 1989||Nov 8, 1995||Apple Computer, Inc.||Vertical filtering method for raster scanner display|
|EP0685829A1||Dec 8, 1989||Dec 6, 1995||Apple Computer, Inc.||Vertical filtering method for raster scanner display|
|WO1986004204A1 *||Jan 7, 1986||Jul 17, 1986||Tech 21 Pty Limited||Imaging method and apparatus|
|WO1990007767A1 *||Dec 8, 1989||Jul 12, 1990||Apple Computer||Vertical filtering apparatus for raster scanned display|
|U.S. Classification||358/448, 345/643|
|International Classification||H04N5/68, G06T5/20, G01S15/89, G09G5/36, G09G1/00, G09G5/395, G09G1/16, G01S7/52|
|Cooperative Classification||G01S7/5206, G09G5/395, G01S7/52028, G01S15/8993|
|European Classification||G09G5/395, G01S15/89D9, G01S7/52S2B1A, G01S7/52S8B2|
|Nov 7, 1995||AS||Assignment|
Owner name: PALOMAR TECHNOLOGIES CORPORATION, CALIFORNIA
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:HUGHES AIRCRAFT COMPANY;REEL/FRAME:007715/0955
Effective date: 19951006
|Jan 4, 1996||AS||Assignment|
Owner name: CHEMICAL BANK, AS AGENT, NEW YORK
Free format text: ASSIGNMENT FOR SECURITY;ASSIGNOR:PALOMAR TECHNOLOGIES CORPORATION;REEL/FRAME:007603/0481
Effective date: 19951006